Han Liu

Han Liu, Tian Liu, Kai Huang

As urban logistics demand continues to grow, UAV delivery has become a key solution to improve delivery efficiency, reduce traffic congestion, and lower logistics costs. However, to fully leverage the potential of UAV delivery networks, efficient swarm scheduling and management are crucial. In this paper, we propose a real-time scheduling and management system based on the ``Airport-Unloading Station" model, aiming to bridge the gap between high-level scheduling algorithms and low-level execution systems. This system, acting as middleware, accurately translates the requirements from the scheduling layer into specific execution instructions, ensuring that the scheduling algorithms perform effectively in real-world environments. Additionally, we implement three collaborative scheduling schemes involving autonomous ground vehicles (AGVs), unmanned aerial vehicles (UAVs), and ground staff to further optimize overall delivery efficiency. Through extensive experiments, this study demonstrates the rationality and feasibility of the proposed management system, providing practical solution for the commercial application of UAVs delivery in urban. Code: https://github.com/chengji253/UAVDeliverySystem

Jaeyeon Jang, Diego Klabjan, Han Liu et al.

This paper proposes a novel multi-agent reinforcement learning (MARL) method to learn multiple coordinated agents under directed acyclic graph (DAG) constraints. Unlike existing MARL approaches, our method explicitly exploits the DAG structure between agents to achieve more effective learning performance. Theoretically, we propose a novel surrogate value function based on a MARL model with synthetic rewards (MARLM-SR) and prove that it serves as a lower bound of the optimal value function. Computationally, we propose a practical training algorithm that exploits new notion of leader agent and reward generator and distributor agent to guide the decomposed follower agents to better explore the parameter space in environments with DAG constraints. Empirically, we exploit four DAG environments including a real-world scheduling for one of Intel's high volume packaging and test factory to benchmark our methods and show it outperforms the other non-DAG approaches.

Han Liu, Pedro Antonino, Zhiqiang Yang et al.

We develop the concept of Trusted and Confidential Program Analysis (TCPA) which enables program certification to be used where previously there was insufficient trust. Imagine a scenario where a producer may not be trusted to certify its own software (perhaps by a foreign regulator), and the producer is unwilling to release its sources and detailed design to any external body. We present a protocol that can, using trusted computing based on encrypted sources, create certification via which all can trust the delivered object code without revealing the unencrypted sources to any party. Furthermore, we describe a realization of TCPA with trusted execution environments (TEE) that enables general and efficient computation. We have implemented the TCPA protocol in a system called TCWasm for web assembly architectures. In our evaluation with 33 benchmark cases, TCWasm managed to finish the analysis with relatively slight overheads.

Ari Seff, Alex Beatson, Daniel Suo et al.

Developments in deep generative models have allowed for tractable learning of high-dimensional data distributions. While the employed learning procedures typically assume that training data is drawn i.i.d. from the distribution of interest, it may be desirable to model distinct distributions which are observed sequentially, such as when different classes are encountered over time. Although conditional variations of deep generative models permit multiple distributions to be modeled by a single network in a disentangled fashion, they are susceptible to catastrophic forgetting when the distributions are encountered sequentially. In this paper, we adapt recent work in reducing catastrophic forgetting to the task of training generative adversarial networks on a sequence of distinct distributions, enabling continual generative modeling.